Japan-Korea : Workshop on Physics of Wave Heating and Current Drive, NFRI, Daejon, Korea, Jan. 14-15, 2008 R F &LHRF& ECRF ICRF JT - 60 JT-60 RF group Japan Atomic Energy Agency Design study for JT-60SA ECRF system and the latest results of JT-60U ECRF system Takayuki KOBAYASHI, Shinichi MORIYAMA, Kenji YOKOKURA, Mitsugu SHIMONO, Koichi HASEGAWA, Masayuki SAWAHATA, Sadaaki SUZUKI, Masayuki TERAKADO, Shinichi HIRANAI, Koichi IGARASHI, Fumiaki SATO, Kenji WADA, Takashi SUZUKI, Shinichi SHINOZAKI and Tsuneyuki FUJII Japan Atomic Energy Agency
Contents I. Design study for JT-60SA ECRF system Outline of JT-60SA and the ECRF system Design study of antennas by a numerical code II. The latest results of JT-60U ECRF system System upgrades 1.5 MW for 1 sec oscillation experiment III. Summary
I. Design study for JT-60 SA ECRF system
1-1. Outline of JT-60Super Advanced (JT-60SA) A Combined project of Japanese national project (former JT-60SC or NCT) ITER satellite tokamak project. Plasma Current I p (MA) high-s for DEMO ITER similar 3.5 / 5.5 (Collaboration between EU and Japan) Toroidal Field B t (T) 2.59 / 2.72 Major Radius (m) Minor Radius (m) Elongation, κ 95 Triangularity, δ 95 Aspect Ratio, A Shape Parameter, S Safety Factor q 95 Flattop Duration Heating & CD power NBI ECRF PFC wall load Neutron (year) 3.16 / 3.01 1.02 / 1.14 1.7 / 1. 83 0.33 / 0. 57 3.10 / 2.64 4.0 / 6.7 3.0 / 3.77 100 s 41 MW x 100 s 34 MW 7 MW 10 MW/m 2 4 x 10 21 D 2 main plasma + D 2 beam injection
1-2. Overview of JT-60SA ECRF System Gyrotrons and Power Supplies 110 GHz 4 units : 1 MW / 100 s JT-60U ECRF system will be upgraded 140 GHz 5 units : 1 MW / 100 s It will be newly fabricated! 3 gyrotrons : provided by EU 2 gyrotrons : developed by JA 5 units of power supplies by EU Injection power & pulse duration 3 MW / 100 s @ 110 GHz 4 MW / 100 s @ 140 GHz Torus Hall P1 1.25 x4 P4 P8 P11 2.5 x5 Total 7 MW /100s All components of the transmission lines will Maintenance Space 110 GHz 140 GHz be developed by JA. Reuse / Upgrade New Gyrotron room (4 th floor)
2-1. Key Issues of transmission systems Key Issues!! All Gyrotron systems can be operated at 1 MW output power with 100 s pulse duration. Transmission lines also enable operation of 1 MW for 100 s. I. Cooling systems are required, miter-bends, antennas Maintenance in JT-60SA Torus Hall is limited owing to high neutron fluence. Only by a remote handling technology is acceptable in vacuum vessel. II. High reliability is required, especially in-vessel-components, such as antenna. JT-60SA can produce various plasma shapes with toroidal field of 1.5 2.7T. (ITER like single null, double null ). III. Various positions of the electron cyclotron resonance surface should be covered by the same antenna. Now we are in the phase of basic design study.
2-2. Conceptual Designs of ECRF Antenna Two types of antennas are now under consideration. Linear Motion Antenna Driving Mechanism Straight Moving Mirror Curved Mirror Corrugated Waveguide Cooling Channel 2-D Scan Antenna Rotating mechanism Focusing Mirror (Toroidal angle control) Flexible Cooling Unit Corrugated Waveguide Flat Mirror (Poloidal angle control) Rotating mechanism Straight Moving Rod
3-1. Principle of a Linear Motion Antenna The linear movement of the flat mirror change the location of reflection on the curved mirror, and alters the beam angle. Drive shaft can be used as a rigid water-cooling channel. Bellows or flexible parts can be placed outside VV. Low risk of water leakage, low frequency of maintenance inside VV 58 deg -27deg dmin=0.2m dmax=0.6m φ63.5mm WG Side View 0.48m Cannel for coolant (fixed) Cannel for coolant (linearly movable) Flexible tube Upper View
3-2. First Design and Numerical Analysis First design of a linear motion antenna was done in 2006 based on the geometrical-optics, in which an ideal lens and para-axial beam were assumed*. Port Waveguide end Linearly moved mirror : M1 height : 480 mm width : 480 mm angle : 35.5 degree Curved Mirror (M2) surface : cylindrical curvature : 700 mm x (mm) Plasma Expected injection angle angle : -27 ~ 58 degree ( from port axis) Fixed cylindrical mirror : M2 z (mm) In actual, the cylindrical mirror dose not work as an ideal lens and beam into the curved mirror is not para-axial beam. Large curvature Small curvature To clarify the property of RF propagation, radiation power profiles were evaluated by a numerical code** based on the Huygens-Fresnel formula. y (mm) Similar to a flat mirror on each reflecting position. Is there an undesirable deformation of beam profile? * S. Moriyama et al., Fusion Engineering and Design, Volume 82, Issues 5-14, October 2007, Pages 785-790 ** Y. Tatematsu et al., Japanese Journal of Applied Physics 44, No. 9A, 6791-6795, 2005.
3-3. RF Profile at Resonance Surface Poloidal direction Toroidal direction 40 20 0-20 θ 40 deg 20 deg 1500 mm 0 deg -20 deg Radiation power profiles in this design were evaluated around the resonance surface. Profile is almost bi-gaussian and the significant deformation was not found. Vertical Expansion of profile on the lower side owing to 1) Incident angle into the surface (~ 1/sin θ) This can not be avoided. Injection to lower position from the upper inclined port causes large vertical expansion. 2) Effect of mirror curvature Can it be minimized by adjustments of the mirror curvature? Does the adjustment conflict with wide range incident angle?
3-4. Optimization of the Mirror Curvature x (mm) Mirror shapes having various curvatures fit into the JT-60SA port. z (mm) Geometry of curve mirrors Larger curvature makes narrower profile. In the case of R=1000mm, vertical expansion is relatively small, w vertical /w horizontal ~ 1.3 Upper half range (0 ~ 58 deg) is available Smaller curvature makes wider incident angle. Expected Full range incident angle (-27 ~ 58 deg) is available at R < 700 mm. In that case, w vertical /w horizontal ~ 2.0 Optimization and clarification of the antenna performance will be continued. 1/e radius of electric field [mm] P total - Pspillover on mirrors Transmission Efficiency P total 200 150 100 W vertical W horizontal 50 Expected radius with an ideal lens assumption 0 500 600 700 800 900 10001100 Curvature of cyrindrical mirror [mm] 1.1 1 R 600mm 0.9 R 700mm R 800mm R 900mm R1000mm 0.8-30 -20-10 0 10 20 30 40 50 60 Incident Angle (degree)
4-1. 2D-scan Antenna Converging Mirror Rotating Mechanism Straight Moving Rods for Rotating and Steering Beam angle (from the antenna axis) ECRF Antenna RF Gate Valve 480 +25(58) deg Mirror Driving System Steering Mirror Steering Mechanism Poloidal Sectional View Corrugated Waveguide Casing Bellows Antenna Support -33(0) -43(-10) -60(-27) Torus Window Stabilizing Baffle Plate Supports for the Antenna System 480 ECRF ECRF Vacuum Vessel Cryostat 0 1 2 3 m Stabilizing Baffle Plate Vacuum Vessel Wall Cryostat Wall 0 0.5 1 m Toroidal Sectional View Toroidal and poloidal cross sectional views of the 110 GHz antenna Schematic view of the antenna installed into the NCT upper port The antenna needs water-cooling system in long pulse operation, 1MW x 100 s. How to cool the antenna? Flexible spiral tube? Other flexible materials? An idea of using a carbon sheet is now under consideration.
4-2. First Design of Focusing Mirror The port has relatively wide area compared with that of JT-60U. Good focusing performance is expected. First design of the focusing (ellipsoidal) and the flat mirror was done by using the numerical code. D 800mm fa 160mm 480mm WG (φ63.5mm) 370mm Distance from the flat mirror (L) Calculation parameter. D : position of one of the focus points (fa). 3000 mm 10 fb Long distance from the WG end to the focusing mirror will make good focusing property. On the other hand, the size of the focusing mirror is limited due to the port size. Here, the distance is set to 800 mm. Focusing property will be adjusted by the focus point of ellipsoidal plane which is shown as D.
4-3. Focusing Property 500 1000 1500 2000 2500 L 1/e electric field radius [mm] 100 80 60 40 20 JT-60U antenna 0-200 0 200 400 600 800 1000 500 1000 1500 2000 2500 Distance from flat mirror (L) Good focusing performance! D : Distance between WG-end and one of the focus points (fa) [mm] Narrow beam radius in wide region will be favored in JT-60SA due to various positions of the resonance surface. In this parameter, D ~ 600 mm, beam radius less than 60 mm in wide region! Therefore, this antenna will produce narrow EC driven current profile for various plasma shapes!
5. Summary of Present Antenna Design Two types of antennas were evaluated by the numerical code. Linear motion antenna Advantage Higher reliability of cooling system with wide poloidal injection angle. Disadvantage Relatively wider beam width for lower side. 2-D scan antenna Advantage Good performance of beam focusing in wide region. Disadvantage Require an idea of a flexible cooling structure with high reliability.
II. The latest results of JT-60U ECRF system
1. Status of ECRF system in JT-60U JT-60U Tokamak (cross sectional view of vacuum vessel) #1 #2 Construction : 1998 Full operation : 2001 Antenna A & B Transmission Lines #3 #4 Gyrotrons 1 MW, 5 s at 110 GHz each 4 high power gyrotrons 4 transmission lines (HE11 mode) 2 antennas (launchers) Injected RF energy 10 MJ (2.8 MW x 3.6 s) by 2002 Long Pulse Operation 45 s (0.35 MW) by 2004 Injected Power (MW) 3 Objectives 1 0.3 2002 2006 0.1 1 3 10 30 80 Pulse Duration (s)
2. The latest upgrades of the system in 2007 High power dummy load system was installed in gyrotron room, it enables gyrotron aging up to 1 MW - 40 s (40 MJ). High power dummy load system Tank Load 1.88m Pre-WG Load by GA RF from gyrotron Vacuum Pumping 1.0MW,5s 1.5MW,CW,Att:50% Sliding support Vacuum Pumping EH 11 HE 11 Main-WG Load by GA 1.0MW,CW,Att:75% Vacuum Pumping Si 3 N 4 DC break Some improvements of gyrotron were carried out to achieve the longer pulse and the higher power. Si 3 N 4 DC break was installed instead of alumina. Cavity cooling water flux was increased about 20 %. Some minor modifications.
3. 1.5MW for 1s at 110 GHz by JT-60U gyrotron JT-60U (@110GHz, 2007) Temperature (deg) 200 180 160 140 120 100 80 60 40 20 0 Temperature on 1.5MW 1sec oscillation Boiling temperature at 3-5 atm Cavity RF Collector (middle part) Cooling water for DC break 0 2 4 6 8 10 Time (s) Gyrotron output power (MW) 1.6 1.4 1.2 1 0.8 0.6 JT-60U (@110GHz, ~2006) JAEA for ITER (@170GHz, ~2007) 0.4 0.2 Effective pulse length for present Tokamak Pulse length for ITER 0 experiments 0.1 1 10 100 1000 10000 Pulse length (s) Ref. State-of-the-Art of High Power Gyro-Devices and Free Electron Masers Update 2006 by Dr. M. Thumm Longer pulse trial, such as 2 sec, will be done with detailed temperature measurements of some components. NEXT step: Improvement of the mode converter planned in this year will enable longer pulse test for JT-60SA.
III. Summary Current status of Design study for JT-60SA ECRF system was shown. Until now, Conceptual specification of ECRF system was shown. Two types of antenna have been considered. Numerical studies for RF propagation by a linear motion antenna and a conventional 2-D scan antenna were carried out. Heat transport by a carbon sheet is now under consideration. (not presented today) Those design study will be continued until 2010. The latest results of JT-60U ECRF system was shown. Some improvements of the system were carried out. High power trial was done and 1.5 MW/ 1 sec oscillation was successfully achieved.